|
I'm using the command line curlPHP code:
|
# ? Dec 27, 2017 22:18 |
|
|
# ? Jun 8, 2024 06:14 |
Grump posted:I'm using the command line curl Okay yes that's a weird thing to do. Does anyone actually recommend ever doing that, if you aren't on the command line already? Try using the API instead. PHP code:
|
|
# ? Dec 27, 2017 22:27 |
|
okay cool. Looks like this was the problem: "Error:SSL certificate problem: self signed certificate in certificate chain" I went ahead and added, which fixed it: PHP code:
|
# ? Dec 27, 2017 23:35 |
|
Grump posted:okay cool. Looks like this was the problem: Didn't you already do that by setting the SSL verify to 0?
|
# ? Dec 27, 2017 23:40 |
|
yeah I did, but I'd rather not write an extra line of code for every request I have to make. This isn't a production project, but rather me just learning. Anyway, I just figured it out. Thanks!
|
# ? Dec 27, 2017 23:43 |
|
I'm new to VCS and github but when viewing files on github, is there any way for it to display the "pretty" version? For example: https://github.com/jonathanstowe/WebService-Soundcloud/blob/master/lib/WebService/Soundcloud.pm It has all of the pod markups but its just displaying everything in raw text.
|
# ? Dec 30, 2017 19:03 |
|
Hughmoris posted:I'm new to VCS and github but when viewing files on github, is there any way for it to display the "pretty" version? That file is already syntax highlighted, which is what I would consider "pretty." By "pretty" do you mean formatted? Like the output of your PM file? Github supports that for markdown but I don't think it does for PM.
|
# ? Dec 31, 2017 19:11 |
|
KernelSlanders posted:That file is already syntax highlighted, which is what I would consider "pretty." By "pretty" do you mean formatted? Like the output of your PM file? Github supports that for markdown but I don't think it does for PM. Ahh. Yes, formatted is the word I was looking for and it makes sense now that Github wouldn't format a .pm file.
|
# ? Dec 31, 2017 19:33 |
|
Does Bash have docstrings like Python or an `END` command like AviSynth after which you can write whatever, or what's the idiomatic way to escape a bunch of text in your files?
|
# ? Jan 1, 2018 13:17 |
|
Sadly the accepted way I think is just a series of hashtags per line:code:
code:
|
# ? Jan 1, 2018 13:49 |
ufarn posted:Does Bash have docstrings like Python or an `END` command like AviSynth after which you can write whatever, or what's the idiomatic way to escape a bunch of text in your files? I think you may be confusing the Python multiline string syntax (that uses triple quotes) with the separate Python syntax of docstrings, which are a bare string as the first statement of a function. It just happens that multiline strings are very often used for docstrings too. Anyway, Nude in the post above already shows off the HEREDOC syntax: Bash code:
|
|
# ? Jan 1, 2018 14:09 |
|
I've been thinking about improving software quality and came across QuickCheck, aka, property-based testing. It seems like a neat idea. Hypothesis is a python (and java?) implementation. Just looking for any thoughts about it...downsides, upsides, anything you want to say about it.
|
# ? Jan 4, 2018 15:39 |
|
Thermopyle posted:I've been thinking about improving software quality and came across QuickCheck, aka, property-based testing. It seems like a neat idea. Hypothesis is a python (and java?) implementation. I've never had a chance to use property-based myself, but fellow forums user MononcQc has a neat blog post on it, and is writing a book on the subject (in Erlang, but a lot of the theory should apply to other languages).
|
# ? Jan 4, 2018 20:26 |
|
redleader posted:I've never had a chance to use property-based myself, but fellow forums user MononcQc has a neat blog post on it, and is writing a book on the subject (in Erlang, but a lot of the theory should apply to other languages). That's interesting. This part: quote:The initial state of the system, as a data structure Made me think of Redux...
|
# ? Jan 4, 2018 21:28 |
|
We use it at work, in addition to other tests. The biggest challenge for me has been figuring out how to get people to move from “that’s neat” to “that’s neat, I’m gonna spend some time to learn how to do it myself.” And obviously, coming up with properties to test can be tricky.
|
# ? Jan 5, 2018 14:15 |
|
Thermopyle posted:I've been thinking about improving software quality and came across QuickCheck, aka, property-based testing. It seems like a neat idea. Hypothesis is a python (and java?) implementation. That sounds a lot like what Microsoft's Pex was/is, but I'm not confident it's the same.
|
# ? Jan 5, 2018 21:45 |
|
I do matched betting (UK based, perfectly legal). One part of this has casino offers where you get to play a silly amount of either slots or blackjack. Blackjack has a higher EV, but has the tedium of playing thousands of hands of blackjack. Considering how to play a hand is just following the well-established optimal blackjack strategy it seems like something that should be very easy to make a bot for. Is this something that could be done with a coding language like Autohotkey? My assumption is the biggest challenge will be OCR on the image to determine what cards the player and dealer both have. Past that, it's just a matter of clicking in a certain area to double/hit/stand/split. I've just started to learn Python (almost finished Learn Python 3 the Hard Way) but I guess this is way past my level.
|
# ? Jan 6, 2018 01:25 |
|
Sad Panda posted:I do matched betting (UK based, perfectly legal). One part of this has casino offers where you get to play a silly amount of either slots or blackjack. Blackjack has a higher EV, but has the tedium of playing thousands of hands of blackjack. Considering how to play a hand is just following the well-established optimal blackjack strategy it seems like something that should be very easy to make a bot for. Making a bot to gamble for you is likely illegal. Besides, the EV has to go to the house over thousands of hands...? That said, OCR may or may not be the way to go, you could also intercept the process memory in some manner to read the variables directly.
|
# ? Jan 6, 2018 02:50 |
|
I'd assume online blackjack would reshuffle after every hand and eliminate the edge from counting cards.
|
# ? Jan 6, 2018 03:16 |
|
It's not counting cards. The +EV is coming from the casino offers, not grinding out the games. AHK has all the functionality you would need to automate this. You know exactly where the cards show up, you can match images against those positions, then feed in the clicks. Wholly doable. That said, you're not the first person to figure out this angle. It would not surprise me if trying to automate it would be illegal, against the terms of service, or otherwise not safe to execute long-term. Online poker games used to take extreme measures to prevent this sort of thing, intentionally obscuring their input to text-drawing syscalls and making pattern-matching difficult. Even if you're clever enough to deal with making the bot "look" like a person (randomize click location, randomize delays between clicks, go off-chart once an hour, etc.) there's still the threat of an administrator opening a private chat and seeing you continue to 'play' while ignoring them.
|
# ? Jan 6, 2018 03:56 |
|
There's also the issue of, how much is your time worth? Is there something actually productive you could be working on instead of writing a bot to make marginal amounts of money gambling?
|
# ? Jan 6, 2018 04:20 |
|
JawnV6 posted:...That said, you're not the first person to figure out this angle. ... This is the main point. It's doable, but only from a "is this possible?" point of view. It might be illegal, so the answer is "yes, but probably not worth it".
|
# ? Jan 6, 2018 05:24 |
|
If you find beer money a useful motivation to learning programming, there's the tried-and-true path of making a video-game bot. Definitely not illegal (though definitely bannable.)
|
# ? Jan 6, 2018 05:38 |
|
Dont see why it would be illegal. Blackjack has solid rules that are common knowledge on what you must do for maximum win chance. Its always hit on less than 17 or something like that. Just taking the hassle out of doing it yourself.
|
# ? Jan 6, 2018 06:37 |
|
NihilCredo posted:If you find beer money a useful motivation to learning programming, there's the tried-and-true path of making a video-game bot. Definitely not illegal (though definitely bannable.) i wouldn’t say definitely
|
# ? Jan 6, 2018 06:42 |
|
Seaside Loafer posted:Dont see why it would be illegal. Blackjack has solid rules that are common knowledge on what you must do for maximum win chance. Its always hit on less than 17 or something like that. Just taking the hassle out of doing it yourself. You don't see why the site would want to ban people from botting? Why do you think they give people those sweetheart deals in the first place?
|
# ? Jan 6, 2018 06:42 |
|
Jabor posted:You don't see why the site would want to ban people from botting?
|
# ? Jan 6, 2018 07:01 |
|
Those people were selling their bots. I think the rules are quite more relaxed if you use it just for yourself from a legal point of view. But I wouldn't try it though.
|
# ? Jan 6, 2018 07:08 |
|
The HonorBuddy ruling isn't particularly applicable to something which isn't distributed to other people, especially if you stop once banned. The damages they successfully argued for were the negative impacts of every single player in a game being a bot plus the time they had to invest into building better anti-botting technology to detect it.
|
# ? Jan 6, 2018 08:18 |
|
Dr. Stab posted:I'd assume online blackjack would reshuffle after every hand and eliminate the edge from counting cards. Nothing as complex as card counting. Simply OCR to save me having to manually click through 1000s of hands of BJ. JawnV6 posted:That said, you're not the first person to figure out this angle. It would not surprise me if trying to automate it would be illegal, against the terms of service, or otherwise not safe to execute long-term. Online poker games used to take extreme measures to prevent this sort of thing, intentionally obscuring their input to text-drawing syscalls and making pattern-matching difficult. Even if you're clever enough to deal with making the bot "look" like a person (randomize click location, randomize delays between clicks, go off-chart once an hour, etc.) there's still the threat of an administrator opening a private chat and seeing you continue to 'play' while ignoring them. Definitely not thinking I'm the first person. Surprised I didn't find anything with a Google. Mainly seems to bring up some game where you pickpocket a bandit. I accept it's quite probably in a grey area, as matched betting in general is. I understand where you're coming from with the private chat angle, but that never happens on betting sites. It did when I used to macro in UO/FFXI however. NihilCredo posted:If you find beer money a useful motivation to learning programming, there's the tried-and-true path of making a video-game bot. Definitely not illegal (though definitely bannable.) Definitely. I'm trying to learn Python anyway, so this could be a project to learn with. Googling around I found https://code.tutsplus.com/tutorials/how-to-build-a-python-bot-that-can-play-web-games--active-11117 although I'm on OS X so would rather make it work without having to load up Parallels.
|
# ? Jan 6, 2018 13:17 |
|
Sad Panda posted:Is this something that could be done with a coding language like Autohotkey? My assumption is the biggest challenge will be OCR on the image to determine what cards the player and dealer both have. Past that, it's just a matter of clicking in a certain area to double/hit/stand/split. I did this exact same thing years ago using C# and everything turned out ok, and I'm in the US. I learned how to make image classifiers too. AHK is probably want you want to do though.
|
# ? Jan 6, 2018 14:59 |
|
Seaside Loafer posted:Might be against the sites terms and conditions but illegal is just silly. Everyone knows the blackjack rules. Lookit this guy thinking there are no silly laws. What a maroon.
|
# ? Jan 8, 2018 14:44 |
|
Someone intimately familiar with Qt and specifically the QTabWidget and its QTabBar? I'd like to shift the tabs a bit to the side for visual reasons, but the only way this works is using the "left" stylesheet attribute, like pretty much any tutorial says. But when I do that, a corner widget will overlap the controls, when things start getting cramped (say resizing window very small). It's like it shifts the tab buttons but not the layout rect. if I try to get margins, borders or paddings on the QTabBar, it gets flat out ignored. Essentially this is my problem: If I remove the shift via left-attribute, the overlap goes away. Halp?
|
# ? Jan 9, 2018 18:47 |
|
Update: I changed this line: set newFile to POSIX file path1 to: set newFile to (path1 as POSIX file) I don't know why, but it works Original question: Stupid AppleScript headache: I'm trying to get a variable to work in conjunction with the file path. I'm using "titleVal" with path1. titleVal has been verified to be a string. When I use it, or any other string var, it skips over the rest of the code. No error. It's in a loop, so it continues the loop. If I have no variable in there, everything works fine. I've tried "quoted form of titleVal" with still the same problem. code:
LP0 ON FIRE fucked around with this message at 19:52 on Jan 10, 2018 |
# ? Jan 10, 2018 17:46 |
|
I work in a Ruby on Rails shop, and we use a local GitLab instance for issue tracking, merge requests, etc. We've been doing code reviews by assigning merge requests to other developers, who look over the changes, comment on them, assign them back, etc. It's not a great system, for a few reasons, and I'm wondering what recommendations people have for better code review tools. This could be either a GitLab add-in, or something completely different that we integrate with our git repository in some way. It's even possible we could abandon GitLab entirely if some other suite works as well for issue tracking, and better for code reviews. It doesn't have to be free, though that would be nice (and would need at least a free trial period to check out).
|
# ? Jan 11, 2018 16:22 |
|
Peristalsis posted:I work in a Ruby on Rails shop, and we use a local GitLab instance for issue tracking, merge requests, etc. We've been doing code reviews by assigning merge requests to other developers, who look over the changes, comment on them, assign them back, etc. It's not a great system, for a few reasons, and I'm wondering what recommendations people have for better code review tools. This could be either a GitLab add-in, or something completely different that we integrate with our git repository in some way. It's even possible we could abandon GitLab entirely if some other suite works as well for issue tracking, and better for code reviews. It doesn't have to be free, though that would be nice (and would need at least a free trial period to check out). GitHub?
|
# ? Jan 11, 2018 16:26 |
|
UpSource looks cool and JetBrains IDEs have great code analysis features
|
# ? Jan 11, 2018 16:34 |
|
Are software requirements specifications documents generally considered "secret" or "protected" information? I'm starting to work on putting one together and I've found a lot of templates / samples / guides, but I'd like to at least take a look at one that was actually used. (I'm actually looking for the SRS for a specific piece of software that's open source, but the people involved with it are in the middle of a major upgrade and I'd prefer not to get on their bad side at a time like this).
|
# ? Jan 11, 2018 16:50 |
|
GitHub’s code review tool is pretty solid now that they’ve added the ability to batch comments. It’s about 80% of what I’d want from an ideal tool and definitely good enough for typical use. I use Phabricator on a different project, and while the code-review tool is theoretically slightly better, it’s also less stable (it loves to wedge itself and force a reload if you do a lot of comment-editing) and doesn’t handle stale comments as elegantly. On the other, its email interactions are quite a bit better, if that’s something you care about. I use BitBucket on yet another project. It’s... acceptable, a lot like GitHub but without many of the little improvements, most significantly batching comments.
|
# ? Jan 11, 2018 17:04 |
|
|
# ? Jun 8, 2024 06:14 |
|
leper khan posted:GitHub? rt4 posted:UpSource looks cool and JetBrains IDEs have great code analysis features rjmccall posted:GitHubs code review tool is pretty solid now that theyve added the ability to batch comments. Its about 80% of what Id want from an ideal tool and definitely good enough for typical use. Thanks for the suggestions - it turns out that at least one of the other programmers is pretty frustrated with GitLab's code reviews, too. I'll take a look at GitHub for sure, and maybe one or two of the other suggestions. I think my supervisor is kind of interested in Jira, so I should probably look into its capabilities so that I know what I'm dealing with if he tries to convince everybody that it's the best thing ever. --- Beef Hardcheese posted:Are software requirements specifications documents generally considered "secret" or "protected" information? I'm starting to work on putting one together and I've found a lot of templates / samples / guides, but I'd like to at least take a look at one that was actually used. (I'm actually looking for the SRS for a specific piece of software that's open source, but the people involved with it are in the middle of a major upgrade and I'd prefer not to get on their bad side at a time like this). In general, I'd assume so. That said, if you're doing this for your employer, they should be able to provide you with some older examples of what they're looking for, and asking for some is a very reasonable request for you to make. My experience is that they vary quite widely in style and quality even within an organization, so I'd also suggest that you not make a mountain out of a molehill. If it comes down to it, just do your best and go with it. Very few places will fire you because someone thinks your first spec document should have been better.
|
# ? Jan 11, 2018 18:07 |